The ABO blood group system is critically important in blood transfusion. The impact of ABO incompatibility on the outcome of hematopoietic transplantation is less well appreciated. In particular, nonmyeloablative conditioning regimens and peripheral blood stem cell grafts, which are the subject of intense investigation in several NIH protocols, may increase the risk of severe hemolytic complications due to ABO incompatibility between donor and host. Massive immune hemolysis caused by "passenger lymphocytes" in the stem cell graft, and pure red cell aplasia (PRCA) due to recipient anti-donor red cell isohemagglutinins, have occurred both more frequently and with greater clinical severity than that seen with myeloablative conditioning regimens and marrow-derived grafts. As a result of these events, the Department of Transfusion Medicine established procedures to monitor, evaluate, and implement management strategies for care of patients receiving ABO mismatched hematopoietic stem cell transplants at NIH. Serial daily laboratory testing, including complete blood counts, chemistries, direct antibody tests (DAT) and other red cell serologic assays were obtained in all lymphocyte-replete minor-ABO incompatible PBSC transplants. Significant hemolysis was observed in 5 of the first 40 patients monitored in this fashion. Since inclusion of an antiproliferative agent in the GVHD prophylaxis regimen may affect the occurrence of hemolysis in this setting, transplant protocols which included an antiproliferative agent such as mycophenolate mofetil (MMF) or methotrexate (MTX) were compared with those in which only a calcineurin inhibitor (cyclosporine (CsA)) was used to prevent acute GVHD. Fifteen patients received CsA alone, 18 received CsA plus MMF and 7 received CsA plus MTX. A significant but modest reduction in posttransplant serologic events was seen in patients receiving CsA plus MTX versus the CsA alone group. This was manifest as weaker reactions involving donor ABO isohemagglutinins against recipient red cells, consistent with inhibition of isohemagglutinin production by MTX. Serologic detection of a positive DAT was poorly predictive of hemolysis; most patients who developed a positive DAT did not have hemolysis, and one patient had a negative DAT at the onset of hemolysis. To evaluate the effect of major ABO incompatibility on donor red cell engraftment following nonmyeloablative stem cell transplants (SCT), we compared transplant outcomes in patients receiving major ABO incompatible nonmyeloablative SCT (fludarabine/cyclophosphamide conditioning) with subjects receiving myeloablative SCT (cyclophosphamide/high-dose TBI). Donor red cell chimerism (detection of donor red cells in the recipient?s blood) was markedly delayed following nonmyeloablative versus myeloablative SCT, median 114 versus 40 days, and correlated strongly with decreasing host anti-donor isohemagglutinin levels. Anti-donor isohemagglutinins declined to clinically insignificant levels more slowly following nonmyeloablative than myeloablative SCT (median 83 versus 44 days). Donor RBC chimerism was delayed more than 100 days in 9 of 14 (64%) and PRCA occurred in 4 of 14 (29%) patients following nonmyeloablative SCT, while neither event occurred in 12 patients following myeloablative SCT. PRCA lasted 123 to 220 days, and patients with PRCA required a mean of 27 red cell units in the absence of other reasons for transfusion support. Conversion to full donor myeloid chimerism following nonmyeloablative SCT occurred significantly sooner in cases with, compared to those without, PRCA (30 versus 98 days). Patients with delayed onset of donor red cell chimerism who did not develop PRCA had a delayed conversion to full donor myeloid chimerism, and were protected from red cell aplasia by a brridge of autologous erythropoiesis. Cyclosporine withdrawal appeared to induce graft-mediated immune effects against recipient isohemagglutinin-producing cells, resulting in decreased anti-donor isohemagglutinin levels and resolution of PRCA following nonmyeloablative SCT. Sudies currently in progress are directed at determining whether clinical and serologic events following major ABO mismatched PBSC transplants are due to differences in the timing of plasma cell, B lymphocyte and myeloid chimerism.